Microsoft Turing Team
Scale deep learning efforts at Microsoft to solve customer and business problems across various products, starting with Search.
Use deep learning models and vector search to understand the intent behind billions of web searches and find relevant documents among billions of web pages.
Find the most relevant documents to user’s search by ranking them.
QnA & Image QnA
Use state-of-the-art machine reaching comprehension models to go through many documents and find answers (either in text or image format) to questions users may have.
People Also Ask
Predict next probable questions that users might ask to pique their curiosity .
Show the most relevant portion of the document to help users quickly assess how relevant the document is to their search.